RE: Quick Guide to Publishing a Thesaurus on the Semantic Web

Yes, I share Ron's doubts about the reality of the original first
sentence and agree with the thrust of his rearrangement. The key thing
is not the manner in which the thesaurus is stored; rather it is the
format of  the reports/outputs you can get out of it. In my experience,
you should expect to be able to get either a text file or an XML file,
or both, out of any respectable thesaurus management system.
Stella

*****************************************************
Stella Dextre Clarke
Information Consultant
Luke House, West Hendred, Wantage, Oxon, OX12 8RR, UK
Tel: 01235-833-298
Fax: 01235-863-298
SDClarke@LukeHouse.demon.co.uk
*****************************************************



-----Original Message-----
From: public-esw-thes-request@w3.org
[mailto:public-esw-thes-request@w3.org] On Behalf Of Ron Davies
Sent: 03 May 2005 18:12
To: Miles, AJ (Alistair); Mark van Assem; public-esw-thes@w3.org
Subject: RE: Quick Guide to Publishing a Thesaurus on the Semantic Web


Mark and Alistair, 

At 17:24 3/05/2005, Miles, AJ \(Alistair\) wrote:



Most thesauri are stored in a relational database, XML file(s) or text
file(s). 
It is usually possible to create an RDF representation of a thesaurus
from its current representation format via some sort of automated
procedure (e.g. a text parsing program, an XSLT transformation etc.)  


I have no objection to the rest of the text which sounds very sensible,
and gives a good example. However I wonder if it is really true that
there are very many thesauri stored in XML files or (really!?!) text
files. From the user point of view the largest category would probably
be thesauri that are stored in thesaurus management systems; whether the
RDBMS that underlies most of those would be visible or accessible to
most users is another question. How about the following text instead?

<quote>
Most thesauri are managed in a automated system, often in a relational
database structure but possibly in other formats. It should usually be
possible to create an RDF representation of the thesaurus, either
directly from the source data or indirectly from standard output of the
thesaurus management system, via some sort of automated procedure (e.g.
a database report, a text parsing program or an XSLT transformation). 
</quote>

I think this might this be a little closer to reality. (Or at least
reality as I perceive it. ;-)

Ron


Ron Davies
Information and documentation systems consultant
Av. Baden-Powell 1  Bte 2, 1200 Brussels, Belgium       
Email:  ron(at)rondavies.be
Tel:    +32 (0)2 770 33 51
GSM:    +32 (0)484 502 393